Codexqa Code Wiki logo

Codexqa Code Wiki

Organization
openqa-cn
codexqa-code-wiki

Builds a local architecture wiki for a repository from the CodexQA symbol graph (no model needed): modules, who calls whom and how often, reading paths, and one self-contained HTML page. Use when the user mentions codexqa-code-wiki, code-wiki, wiki, knowledge graph, architecture wiki, module map, reading guide, 代码知识图谱, 架构 Wiki, 模块地图, 阅读导览, wiki inputs, --no-llm, community detection, 社区检测, HTML 报告, or asks to map modules / generate a repo wiki without an LLM. Not change review (that is codexqa-code-analyzer), not CodexQA evidence-pack HTML review (that is codexqa-code-reviewer), and not SAST+agent code-risk scan reports (that is codexqa-defect-analyzer). Former skill name: code-wiki.

Overview

Publisheropenqa-cn
Repositorycodexqa
Skill namecodexqa-code-wiki
Stars
233
Forks
3
Bundled files
22
LicenseApache-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 22 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by openqa-cn on GitHub. Read the source before you install it.

Installation

Install the Codexqa Code Wiki AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/openqa-cn/codexqa/tree/main/skills/codexqa-code-wiki
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/openqa-cn/codexqa.git /tmp/codexqa
mkdir -p .claude/skills
cp -r /tmp/codexqa/skills/codexqa-code-wiki .claude/skills/codexqa-code-wiki
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Codexqa Code Wiki in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Codexqa Code Wiki on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Codexqa Code Wiki is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Code Wiki

Three commands, one file to write:

text
node <skill>/scripts/wiki.mjs brief <repo>   # index + export facts + print the brief
  → you write .codexqa-wiki/<repo>/notes.json  # names and sentences only
node <skill>/scripts/wiki.mjs build <repo>   # validate + render one HTML file

The script owns every fact on the page: modules, sizes, call counts, groups, the architecture diagram, public symbols, files, links, search. You own the words: what the repo is, what each module does, and which reading path fits which task. You never edit HTML, CSS, or diagrams.

<skill> is this skill's directory. <repo> is a local checkout (preferred). README.md / README.zh-CN.md are for humans; do not load them.

Documents (load on demand)

FileLoad when
SKILL.md (this file)always
references/notes.mdbefore writing notes.json
references/playbook.mdone module only, index problems, persisting for the Web UI
references/cli.mdcodexqa missing, PATH, repo ids, maintenance

Default flow

  1. Brief. Run node <skill>/scripts/wiki.mjs brief <repo>. It indexes incrementally, exports codexqa wiki inputs, and writes .codexqa-wiki/<repo>/ with inputs.json, meta.json, brief.md, and a notes.json scaffold (kept if it already exists). Read the printed brief: groups, suggested paths, and per module its files, calls, callers, public symbols, and key flows.
  2. Notes. Read references/notes.md, then fill notes.json. Every P01-style id and group id in it must exist in the brief. Write 简体中文 unless the user asked for English ("lang": "en").
  3. Build. Run node <skill>/scripts/wiki.mjs build <repo>. If it prints error: lines, fix those fields and rerun; they name the field and the fix. check <repo> validates without writing HTML.
  4. Deliver. Tell the user the printed file path plus two or three lines on what the wiki says. Do not paste the HTML or open it in chat.

The user only wants a quick look → build <repo> --draft right after brief. The page uses rule-based names and says it is a draft. Offer to write the notes afterwards.

Useful flags: brief --limit 8 (large repos), brief --lang en, brief --skip-index (index is fresh), build --out FILE, build --source-url 'https://host/repo/blob/{commit}/{path}' (file links when the remote is not GitHub / GitLab / Gitee / Bitbucket).

Scenario routing

User is asking…Do
What is this repo / module map / architecture wiki / knowledge graphDefault flow
Where should I start reading / onboarding pathDefault flow; put the effort into guides
What does module X do / who does it talk toAnswer from brief.md; deeper digest in playbook Explain one module
Store a rule-only wiki for the CodexQA Web UIplaybook Persist rule-only wiki
brief failed / empty export / wrong branchplaybook Index health

Change review, callers of a symbol, test gaps, and stack traces belong to codexqa-code-analyzer.

Rules

  • Facts come only from this run's export. Do not add modules, calls, or responsibilities that the brief does not support; do not use the README or a website as evidence for a module's role.
  • A reading path follows real calls: each step must call or be called by the next. build rejects anything else and lists the valid neighbours.
  • Standalone modules (no calls to or from other modules) get a name and a role, never a layer.
  • Write for someone who just cloned the repo. No internal field names (deps, cross_community, called_by, wiki inputs); build rejects them.
  • Do not run codexqa wiki without --no-llm, wiki embed, or query wiki. Those call a model or need model-written pages.
  • Do not hand-edit the generated HTML. Change notes.json and rebuild.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Codexqa Code Wiki AI skill do?

Builds a local architecture wiki for a repository from the CodexQA symbol graph (no model needed): modules, who calls whom and how often, reading paths, and one self-contained HTML page. Use when the user mentions codexqa-code-wiki, code-wiki, wiki, knowledge graph, architecture wiki, module map, reading guide, 代码知识图谱, 架构 Wiki, 模块地图, 阅读导览, wiki inputs, --no-llm, community detection, 社区检测, HTML 报告, or asks to map modules / generate a repo wiki without an LLM. Not change review (that is codexqa-code-analyzer), not CodexQA evidence-pack HTML review (that is codexqa-code-reviewer), and not SAST...

Why use Codexqa Code Wiki on TypingMind?

Because you install it once and use it with any model. Codexqa Code Wiki is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Codexqa Code Wiki in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/openqa-cn/codexqa/tree/main/skills/codexqa-code-wiki. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Codexqa Code Wiki?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Codexqa Code Wiki?

As many as you like. As long as a model supports skills, you can use Codexqa Code Wiki with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Codexqa Code Wiki AI skill free?

Yes. It is published on GitHub by openqa-cn under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇